Kinematic Comparison Movement
MR Imaging ofPassive Techniques’
ofthe Patellofemoral Positioning and Active
the steady
Frank G. Shellock, PhD Jerrold H. Mink, MD Andrew L. Deutsch, MD Thomas K. F. Foo, PhD
quence
state
(GE
kee)
has been
cient
temporal
kinematic patients underwent magnetic resonance (MR) imaging of the patellofemoral joint in an evaluation of passive positioning and active movement
Thirteen
kinematic
MR
imaging
techniques.
Six-
teen joints were symptomatic, and 10 were not. Delineation of normal and abnormal patellar alignment and tracking was similar with the two techniques, but kinematic MR imaging performed with active movement allowed a substantial reduction in examination time while permitting evaluation of the contribution
of associated
activated
muscles and soft-tissue structures patellofemoral joint function. Index Knee,
terms: Knee, MR. 453.1214 cine study
(MR),
Radiology
1992;
abnormalities, Magnetic
184:574-577
of the patellofemoral joint related to incongruency between the patella and femoral trochlear groove are the frequent cause of patellar subluxation, patellar dislocation, chondromalacia, and arthrosis (1-5). magnetic
imaging
resonance
is a sensitive
(MR)
method
(6-14).
spoiled
Recently,
an ultrafast,
gradient-recalled
acquisition
in
passive achieved
positioning with the
Tower
bly.
Received
January
quested
March
accepted
March
F.G.S. C
574
RSNA,
1992
#{149} Radiology
described Medical transmit
and
Study
that
with
of the
msec, patients
a provisional
clinical
patellar
alignment
abnormal
were
evaluated.
increments
of
and
time
time
onds.
The
of joint
total
image
was approximately was
26 years). Patelhad been per-
three
different
entire
acquisition
flexion,
locations
excursion
of the
to evaluate
lateral retinacular releases, i = 4). Positioning device-A nonfenromag-
kinematic
active
imaging this
studies
device
restrict joints was positioning
(6-8,10,14).
is crucial
give
to displace
the
an erroneous
The
patient
lower
was placed
extremity
tamed
(ie, the
patient
was
between
the
on the
main-
observed
position
and
patient’s
the
feet
procedure
was
was
performed
receive
coil
quadra-
of a i.5-T
spoiled
64-MHz
Systems)
GRASS
previously
described
imaging
6.5/2.1; averages,
of view,
flip
parameters
38 cm; section
movements
of the lower
instructed to initiate active of the joints approximately after hearing the gradient
Systems,
RSNA 21, 1992;
5; revision 27. Address
Milwaukee scientific revision
received reprint
assemreMarch 16; requests to
occur
during
ral joints
flexion
(7,8,10),
because
excessive
rotation
may
abnormal
which internal
be partially
patellar
extremities
of the
alignment
to
patellofemo-
is important
then
or external
responsible and
movement with the
extended
rotational
for track-
thickness,
7
to perform this type of MR imaging with a fast low-angle shot pulse sequence (eg, turboFLASH; Siemens, Iselin, NJ). For kinematic MR imaging with the
scheme
allows
were
mm; acquisition time, 8 seconds for six images at a single section location (14). Alternatively, it would also be possible
positioning
Los Angeles
4X
angle, 30#{176}; number of one; matrix, 160 x 128;
90048 (F.G.S., J.H.M.,
Vicen-
GE
is required for MR imaging.)
Medicine,
444 5 San
an
sequence.
[14],
and software spoiled GRASS
MR
and
pulse
active began
Center,
(ie,
and
(GE Medical
hardware ultrafast
field
care to so that
was
imager
used: signal
of patel-
alignment
body
The following
unpub-
prone
groove Total examipositioning,
imaging
the transmit
tune
(As or
and
with special extremities
in an upright
of
may occur position is
appearance
device the lower
use
patella
with
ultrafast
anterior
lar subluxation (Shellock FG, lished observations, 1987). positioning position
MR
The
because
lateral direct pressure that with the patient in a prone sufficient
technique-MR
would
kinematic
imaging
the
in rela-
patient positioning, patient instructions, etc) was approximately 15 minutes. Kinematic MR imaging: active movement
device with a cut-out uninhibited movement or forces acting on the
movement
MR
it
in
patel!a
tion to the femoral trochlear during joint flexion (6-10). nation time for the passive
patellofemoral joints that motion) of the patellofemoral used for both the passive
time
and
sections
formed in five of the 16 symptomatic patellofemoral joints (medial transposition of the extensor mechanism, n = 1;
netic positioning area to permit (ie, no pressure
34 sec-
acquisition
to obtain
half;
38 cm;
4 minutes,
necessary
(500/
one
of view,
5 mm;
per degree
500
20 msec
averages,
x 128; field
thickness,
time
flexion,
(TR),
(TE),
of signal
256
section
five
eight women (26 patellofemo16 with symptoms of pateldisorders), and the age range (mean, surgeries
number
matrix,
track-
were
echo
spin-echo, performed at of joint
repetition
and
20);
diagnosis
There
as previously
coil was used. Ti-weighted, axial plane imaging was
Methods
subjects-Thirteen
of the
(6-8,10). A i.5-T 64-MHz (GE Systems) MR imager with a and receive quadrature body
as follows:
Materials
kinematics
maintained along with the relative amount of internal or external rotation of the lower extremities). This patient
Imaging
te Blvd. Los Angeles, CA A.L.D.); UCLA School of (F.G.S.); and GE Medical (T.K.F.F.). From the 1991
method versus active movement
the
at 5#{176} increments,
flexion
each
distance From
evalu-
technique.
while
I
of
to compare
of the patterns of patellar alignand tracking achieved with the
and
of evalu-
ating patellar alignment and tracking, as well as the anatomy of the patellofemoral joint (6-16). Examination of the patellofemoral joint with kinematic MR imaging is typically conducted by using a special device to passively position the joint in 5#{176} increments from extension to 30#{176} of flexion while Ti-weighted, spin-echo images are obtained at each of these positions
ation ment
patel-
The purpose
was
Also,
ing technique-The positioning device was used to passively move the pate!lofemoral joints from extension to 30#{176} of
performed
of the
(1,3,4).
patellofemoral joint are the same with prone and supine positioning (10). Kinematic MR imaging: passive position-
has suffi-
to be
(10,14).
was 14-37 years lar realignment
BNORMALITIES
Kinematic
imaging
joint
ing
to permit
movement
investigation
men and ral joints, lofemoral
453.421 resonance
#{149}
this
seMilwau-
that
resolution
active
lofemoral
pulse
Systems,
developed
MR
during
ing
to
(GRASS)
Medical
Joint:
technique, patellofemoral
positions.
to simultaneously
the
The patient
patient joints in
was
movement 1 second noise and
flex both
pate!-
lofemoral joints until the bottoms of their feet touched the top of the inside of the MR imaging system, and to leave
August
1992
b.
e.
f.
C.
Figure
were
1.
Axial
kinematic
obtained
MR
at the same
the patellae that trochlear grooves in a and b).
images
section
(500/20)
location
improves (ie, the patelbae are shallow, and a slight
them
in this
noise ticed
stopped (14). The patient practhis maneuver two or three times
without
imaging
movement tion ager
position
to ensure
was
was even operator
until
constant)
evenly
at
spaced
move into redundancy
the
(ie, that
mo-
to the
im-
through
the
described six images
that range
were of mo-
tion achievable within the bore of the magnet, and this procedure was repeated for three different section locations length joints
(14). Depending of the patient, could be flexed
within
the confines
system
(14).
The total image the aforementioned Volume
184
on the lower leg the patellofemoral up to 40#{176} or 45#{176}
of the MR imaging acquisition time with MR imaging pa-
Number
#{149}
2
joints
more centralized of the lateral
rameters
in a 23-year-old
was
positions) retinacula
approximately
and
woman,
Total
examination
during is evident
for the
ac-
tive movement kinematic MR imaging procedure, including time allotted for patient positioning, patient instructions, and practice runs, was approximately 7 minutes. Display, evaluation, and comparison of the kinematic MR imaging studies.-Images obtained during kinematic MR imaging with the passive positioning and active
movement
uated
as single-frame,
techniques
static
were
images
with
There
eval-
and
also were displayed as a cine loop (68,14,15). Compared with individual static images, evaluation of patellofemo-
ral joint
passive
is bilateral
the higher increments at the early increments
40 seconds
time
obtained
(f) 30#{176} of flexion.
(ie, one series to obtain images at six section locations during extension and three series to obtain six images at three section locations for thorough eva!uation of patellar alignment and tracking) (i4).
the previously rate of movement provided at a given section location Imaging
patellofemoral
at (a) 5#{176}, (b) 10#{176}, (c) 15#{176}, (d) 20#{176}, (e) 25,
gradient
that
acceptable and (14).
the
of the
positioning.
lateral
of joint flexion. of joint flexion
congruency
Images
subluxation
with
of
The femoral (arrowheads
the cine-loop
display is considered to provide the best qualitative information concerning the various patterns of patellar alignment and tracking (6-8,14,15). In addition, the cine-loop display of the images allows the determination of whether patellofemoral joint incongruency (ie, the position or positions of the relative to the femoral trochlear groove) identified during these studies improved, worsened, or remained the abnormal
patella
same during information
mine
joint flexion is considered
the severity
(6-8,14). to help
This
deter-
of the abnormality
(2,13).
Previously
ria and normal
were joints
described
qualitative
crite-
descriptions of normal and abpatellar alignment and tracking used to assess the patellofemoral (6-8).
Briefly,
normal
patellar
Radiology
575
#{149}
b.
e.
f.
C.
Figure
2.
the same of joint
Axial
MR
section flexion
images
of the
location is evident,
patellofemoral
and
in the same
and
the
joints,
patient
patterns
of the patellofemoral transverse
or lateral This
facets
ion,
causes
“centered”
the
relative
ral trochlear groove. of the patella occurs the patella is laterally
the lateral
facet
laps
the
aspect
lateral
trochlear eral
groove
pressure
acterized functional dominant
With ELPS, subluxation 576
to
femo-
groove femoral of the
(6,7,10).
10).
(ELPS)
lateralization, lateral patellar
groove
femoral
of the patella
is char-
with a
a small amount of lateral may occur during joint flex-
#{149} Radiology
movement
as increasing
taut
tracking
are
similar
tension
from
soft-tissue
displaces
kinematic
of the bilateral to those
one
or
structures
the
patella.
of the patellar
ridge
groove femora!
Lateral-to-medial
relative or the trochlea
subluxation
or centermost
part
Me-
to the center(6,7, is
In addition of patellofemoral qualitative tween the
and
tracking
medially of joint
to the above classification joint abnormalities,
comparison was made bepatterns of patel!ar alignment
shown
with
the passive
technique
(6.5/2.1;
lateral
subluxation
depicted
in Figure
positioning
flip
during
angle,
30#{176}) and
higher
at
increments
1.
method
versus
with
the
ac-
tive movement technique. This comparison was not performed in a “blinded fashion”; the viewing of the images from these two different studies, however, was performed randomly (ie, the evaluation of the images obtained with the passive positioning technique was not selectively performed before evaluation of images obtained with the active movement technique, or vice versa).
Results
of the femo-
ral trochlea and then displaces during the higher increments flexion (6,7,10).
lat-
usually onto facet (6,7,10).
and
overly
active
improvement
displayed when there is a slight-to-moderate lateral subluxation of the patella during the early increments of joint flexion; as flexion continues, the patella moves across the femoral trochlear
over-
Excessive
alignment
femoral trochlear most part of the
or the trochlea
of the patella
syndrome
by tilting
ment
medial
patella
1. The
the
dial subluxation of the patella, or patella adentro, is shown by medial displace-
(6,7,10,16). to the
with
progressively
Lateral subluxation when the ridge of displaced relative
to the femoral trochlear centermost part of the
and
more
without of the
of the patella
orientation
appear
joint,
displacement
as in Figure
of patellar
alignment and tracking is displayed with the ridge of the patella positioned in the femoral trochlear groove as it travels in a vertical plane during flexion
obtained
Technically
acceptable
images
were
obtained in all patients examined investigation. Ten patellofemoral were normal, and these were the 10 joints
that
were
asymptomatic.
in this joints same Of
the pate!lofemoral joints that were symptomatic, six had lateral subluxation of the had
pate!la, medial
two subluxation
had
ELPS,
and
of the
eight
patel!a.
August
1992
None of the patellofemoraljoints lateral-to-medial subluxation
tella.
showed of the pa-
Of the five patellofemoraljoints
which
a surgical
realignment
ments,
in procedure
including
should
had previously been performed, one had lateral subluxation of the patella and four had medial subluxation of the patella. Delineation of each of the forms of patellar alignment and tracking patterns with the passive positioning technique (Fig 1) was comparable with that achieved with active movement kinematic MR imaging (Fig 2), such that norma! patellar alignment and tracking, lateral subluxation of the patella, ELPS, and media! subluxation of the patella were similarly shown with each method for each patellofemoral joint.
be noted
tative
criteria
passive
Abnormal
conditions
of
ment and tracking begin liest portion of the range this joint, as the patella ulates with the femoral (2,7,i3,i6,i7).
creases,
As
the
enters and antictrochlear groove
flexion
patella
patel!ar alignduring the earof motion of
of the
moves
joint
deeper
into
(2,6,7,10,13-15,17). alignment and
shown technique
with the paswere similar
to those shown with the active movement kinematic MR imaging method. The image quality of the kinematic studies
performed
with
the
ultrafast
spoiled GRASS pulse sequence was adequate to show the anatomy of the bone, as
well
as
the
soft-tissue
structures
asso-
dated with the function of the pate!lofemoral joint (14). Previous radiographic studies comparing passive positioning techniques with dynamic imaging methods for assessment
have
of the
been
patellofemoral
performed
son of ultrafast static skyline
show good quantitative
(12,15).
joint
Compari-
CT views with single, views has been reported
correlation between certain patellofemoral measure-
methods relationships
2
and in this
tracking
to compare
to
the
none
of the
matic
MR imaging
with
studies
were
3.
4.
patellofemoral to be completely
observed
passive
with
5.
system
The results of kinematic
(GE
Medical
showed that MR imaging
respect
to the
6.
active-
Systems).
patel-
Acta Orthop
Scand
in the rabbit.
1989; 60:188-191. Shellock FG, Mink
1941; 12:319-
F, Frich LH. by patellar sub-
Acta Orthop
JH, Deutsch
of patellar
Scand
A, Fox JM.
tracking
abnormali-
ties using kinematic MR imaging: clinical experience in 130 patients. Radiology 1989; 172:799-804.
7.
8.
Shellock FG, Mandelbaum B. Kinematic MRI of the joints. In: MinkJH, Deutsch AL, eds. MRI of the musculoskeletal teaching file. New York: Raven, Shellock FG, Mink JH, Deutsch
Ferkel
RD.
system. A 1990. AL, Fox JM,
of patients with after lateral retinacutar release by kinematic magnetic resonance imaging of the patellofemoral joint. persistent
assess-
pate!-
joint during active movement by using ultrafast CT, Hyperscan MR imaging, or ultrafast MR imaging is considered to be advantageous, because the contribution of activated muscles and other soft-tissue structures to patellar alignment and tracking may be evaluated (i2,14,i5,i7). This results in a more physiologic examination of the patellofemora! joint. With respect to the specific use of ultrafast spoiled GRASS MR imaging to study the patellofemora!joint, the main advantages include a substantial reduction in overall examination time (compared with duration in the previously described method) (6), as well as the fact that there is no need for a sophisticated positioning device for passive flexion of the joint. Because the examination time for performance of kinematic MR imaging of the patellofemoral joint during active movement is relatively short, it is feasible to include this procedure along with routine MR imaging examination of the knee, since both procedures can be cornpleted within an acceptab!e time in the clinical setting. This would provide a more thorough, overall examination of the knee, which would be useful in cases of combined abnormalities (1-4) and in cases of suspected patellofernoral incongruency with signs and symptoms that mimic other types of internal derangernents of the knee. U
in recurrent
410. Moller NB, Moller-Larsen Chondromalacia induced
Evaluation
the two forms were similar
qualitative
relationships
M, NeliPatel-
lar dislocation. J Bone Joint Surg [Br] 1989; 71:788-792. Ficat RP, Hungerford DS. Disorders of the patello-femoral joint. Baltimore: Wilhams & Wilkins, 1977. Wiberg G. Roentgenographic and anatomic studies on the femoropatellar joint, with special reference to chondromalacia
luxation
movement kinematic MR imaging studies performed with a Hyperscan MR imaging
tTM, Osterman K, Kormano 0, Hurme M, Taimela S.
patellae.
kine-
with
compared
Kujala markka
lofemoral
active moveCT study, the
that
patterns
2.
quanti-
(i5). Furthermore, it was conthat the cine loop of the ultrafast CT views of patellar movement provided the best qualitative assessment of patellofemora! relationships (15). In a prior study performed by our group (12), the patellar alignment and
9.
Arthroscopy 1990; 6:226-234. Shellock FG, MinkJH, Fox J, Ferkel Friedman M, Molnar T. Kinematic imaging evaluation of symptomatic tients following two or more patellar
alignment 10.
surgeries
1:175. Shellock
JMRI
MR pare1991;
JH, Deutsch A, PressMRI of the joints: techniques and clinical applications. Magn Reson 1991; 7:104-135. Shellock FG, Mink JH, Deutsch AL, Meeks T, Fox J, Molnar T. Axial loaded stress views and kinematic M evaluation of
patellar
12.
(abstr).
R,
FG, Mink
man BD.
11.
Evaluation
symptoms
Kinematic
alignment
and tracking:
results
98 patellofemoral joints (abstr). 1990; 177(P):263. Shellock FG, Cohen MS, Brady
Pfaff JM. and
Evaluation
tracking:
matic MR imaging perscan
T, MinkJH,
of patellar
comparison
alignment
between
and “true”
MR imaging
(abstr).
in
Radiology
kine-
dynamic JMRI
hy-
1991;
1:148-149. 13.
14.
Kujala
UM,
Osterman
K, Kormano
M,
Komu M, Schlenzka D. Patellar motion analyzed by magnetic resonance imaging. Acta Orthop Scand 1989; 60:13-16. Shellock FG, Foo TKF, Deutsch A, MinkJH. Patellofemoral joint: evaluation during acfive
flexion
with
ultrafast
spoiled
GRASS
MR imaging. 15.
16.
Radiology 1991; 180:581-585. W, Phelan J, Kathol MH, et al.
Stanford Patellofemoral joint motion: evaluation by ultrafast computed tomography. Skeletal Radiol 1988; 17:487-492. Nordin M, Frankel VH. Basic biomechanics of the musculoskeletal system. 2nd ed.
Philadelphia: 17.
Lea & Febiger,
1989.
MinkoffJ, Fein L. The role of radiography in the evaluation and treatment of com-
mon anarthrotic lofemoraljoint. 8:203-260.
disorders Clin
of the patel-
Sports
Med
1989;
References Kummel 1980;
#{149} Number
used
of quantitating appears
BM.
lofemoral
184
although
pa-
and It
reliable cluded
1.
Volume
offset,
ment of the normal and abnormal lofemora!joint findings (12). Examination of the patellofemoral
this portion of the range of motion are best suited for identification of instabi!ity of the patellofemora!joint (2-4,6, 7,iO,i3-i5,i7). These imaging techniques include conventional computed tomography (CT), ultrafast CT, and ki-
tracking patterns sive positioning
were
ment techniques authors recognized
with
in-
the femora! trochlear groove. At that point, patellar displacement is less likely to occur, because the femora! trochlear groove functions to buttress and stabilize the patella (3,4,iO,i5,i6). Since patellofemoral incongruency is most likely to exist during the iitia! degrees of flexion of the patellofemoral joint, imaging techniques that show the patella and femoral troch!ear groove during
nematic MR imaging In this study, patellar
tangent
that,
positioning
positioning
Discussion
the
tellar tilt angle, congruence angle, lateral patellofemoral angles (i5).
The
diagnosis
derangements.
of patelPrimary Care
7:199-216.
Radiology
#{149} 577